Love.Law.Robots.

Laptop Shopping!

Featured Imaged `

Here’s something a little light-hearted as I ponder about where to dump my blogging energies next.

I need a laptop!#

Source: https://www.flickr.com/photos/30760976@N04/9605358064

Source: https://www.flickr.com/photos/[email protected]/9605358064

The Problem: I bought my current 2015 Macbook Pro 15″ at the beginning of 2018. I wanted the Mac to be my development machine, where I would do most of my programming. This was way before I decided to try out data science. I didn’t need anything fancy, so I got a laptop which I liked. Aren’t Macs pretty! 😛

Fast forward to the second half of 2020. A pandemic forced me to have time at home to hone my skills. I decided to run 5 million iterations on my development machine. (I would probably cover this in a later post.) The Macbook heaves and hoes. I run the same code on the Surface Pro 7, which is a full kilogram lighter, and it runs the code at the same speed.

My macbook struggles to run code.

This should henceforth be referred to as the silly walks of doom.

Not to mention, the battery here is a bit bloated… not sure what to make of it.

What I need: Unfortunately, times have changed. I need a computer with power. The macOS, unfortunately, has also been irritating me with its change to zsh, breaking compatibility with other programs for their 64-bit requirement, and having asinine security features. This might be the time I get a Linux laptop.

Contender 1: System 76’s Oryx Pro#

Image of System76 Oryx Pro, an option for my laptop.

I have been eyeing System76 products for several years. There’s a lot of things I like about this laptop — tons of options to upgrade, even though the base configuration is fairly maximal already. It’s also got open source written all over it. That’s something I can get behind!

Contender 2: Lambda’s Tensorbook#

Image of Lambda Tensorbook, an option for my laptop.

This laptop does not enjoy as much marketing as our previous entry. Maybe it’s because it does not need it. Its specifications are mouth-watering and I will be expecting that I will never need to look back again after getting it. There’s something to be said about the price (US3,300). However, pushing the Oryx Pro to match these specifications ends up with the TensorBook being cheaper. It’s definitely value for money… as compared to…

Contender 3: Something I can buy in Singapore…#

Picture of a Surface Book. Not so suitable for deep learning but still an option.

A Microsoft Surface Book 2.

It’s disturbing that my preferred choices are located overseas. Unfortunately, my options in Singapore are limited. A computer sold from a major retailer does not have a Linux option. Furthermore, a computer which matches the Tensorbook’s specifications with a Windows or Mac is very expensive. An over S$4,000 laptop for personal use will be one of the most expensive computers to find in Singapore. Maybe everyone is really happy with their options here, or I would just have to get a gaming PC.

Conclusion#

I’m quite hesitant to expend several thousand dollars overseas on a product which might not ship here. Or if it does, the voltage setting for some reason might not be compatible and it will become a dead brick. However, times are changing and retailers relying far more on overseas customers seem to be more acute of our requirements. Is this the time to take a plunge? If you have any experience with these PCs, feel free to let me know!